WO2008040846A1

WO2008040846A1 - Apparatus and method for determining the functional state of a brain

Info

Publication number: WO2008040846A1
Application number: PCT/FI2007/050525
Authority: WO
Inventors: Anu Holm
Original assignee: Työterveyslaitos
Priority date: 2006-10-03
Filing date: 2007-09-27
Publication date: 2008-04-10
Also published as: FI20065627A0; FI20065627A

Abstract

The present publication describes an apparatus and method for determining the functioning state of a person's brain. The apparatus comprises at least one sensor (1, 2), which can be attached to the head (5) in order to measure the functioning state of the brain on the basis of at least two frequency bands (alpha 22, theta 21), signal-processing means (3) for processing the signals measured by a sensor (1, 2), and signal-interpretation means (4) for interpreting the measured signal. According to the invention, the apparatus comprises, in addition, means (4) for detecting from the signal features indicating the functioning state of the brain, on the basis of at least two separate frequency bands (21, 22), as at least a partly continuous process, and means (4) for forming a ratio on the basis of these two frequency bands (21, 22) as at least a partly continuous process.

Description

Apparatus and Method for Determining the Functional State of a Brain

The present invention relates to an apparatus, according to the preamble of Claim 1 , for determining the functioning state of a brain.

The invention also relates to a method, according to the preamble of Claim 13, for determining the functioning state of a brain.

Introduction

According to the prior art, methods based on the measurement of the electrical activity of a brain (EEG) are used in the evaluation of workload (Fairc lough &Venables, 2006; Gevins, et al., 1995; Wilson & Fisher, 1995).

Event-related-potential research

In event-related-potential research, it has been observed that particularly the so-called P300 component changes according to variations in workload (Kok 1997). The use of event-related potentials, (ERP), particularly in field research, is limited by the fact that an external stimulus is required for their measurement, hi a working environment, the presentation of an external stimulus (for example, a beeb sound at one-second intervals) may be impossible to implement, hi addition, some event-related-potential measurements (such as the P300) require a reaction to the presented stimulus. This prevents, among other things, the study of situations causing workload, in which a concrete work task cannot be isolated (Pope, Bogart, & Bartolome, 1995) as its own entity (e.g., thinking about a solution to a demanding problem, writing a scientific article, etc.). Therefore research has been started on methods, which permit the study of workload without the use of an external stimulus and a precise time-synchronization with the actual work task.

The frequency content of EEG

The frequency content of the electric activity of the brain has also been shown to change when the difficulty of the work task changes; theta activity has been shown to increase and alpha activity to decrease as the workload imposed by the task increases (Fairclough & Venables, 2006; Gevins, et al., 1998; Sauseng, Klimesch, Schabus, & Doppelmayr, 2005; Klimesch, 1999). Most of the studies have concentrated on investigating the changes in the posterior parts of the brain, but some have also examined the changes in the frequency content of the EEG measured in the frontal parts of the brain. It has been observed that the increase in theta activity is the largest particularly in the frontal channels of the brain while increasing the load of the working memory task, whereas alpha activity has decreased most of all in the posterior parts of the brain (Gevins, et al., 1998). The time pressure of the task increases the amount of theta activity of the frontal parts of the brain and reduces the alpha activity of the posterior parts (Slobounov, Fukada, Simon, Rearick, & Ray, 2000). hi the research by Slobounov et al., additionally, a larger decrease in alpha activity and an increase in theta activity was observed when the given tasks were answered correctly, compared to situations in which the answers to the tasks were incorrect.

EEG indices

In addition to calculating individual frequency bands, various indices have been derived from the frequency bands, which have been used especially as biofeedback on a person's mental state while investigating adaptive interfaces (Pope, Bogart, & Bartolome, 1995; Scerbo, Freeman, & Mikulka, 2003). Traditionally, EEG-based indices have been calculated as ratios between the various power bands of a specific area of the brain. The traditional indices are obtained as follows: the spectral powers of the measurement points P3, Pz, P4, and Cz are combined and the ratio beta/(alpha+theta), I/alpha, or beta/alpha is calculated. The measurements have mainly concentrated on the measurement of the activity of the posterior parts of the brain, paying less attention to the measurement of the activity of the frontal parts. Several measurement electrodes have been used in the calculation of the indices.

State of the art

In the patent literature, the matter is described in, for instance, US patent 5 460 184, which discloses a method and a device for measuring mental concentration. The method is based on comparing the power of the alpha-frequency range with the power of the entire frequency range.

Correspondingly, US patent 6434419 discloses a multi-parameter method for determining general intelligence, in which, among other things, the theta and alpha bands are measured in connection with neuropsychological and psychometric tests. On the basis of the parameters produced by these three methods (neuropsychological tests, psychometric tests, and neurophysiological variables), a multi-variate model describing general intelligence is formed with the aid of regression or neural-network methods. As an end result, the method produces a single-order number of the person's intelligence. The implementation of the method disclosed is complicated.

Description of the invention

The invention is intended to eliminate the defects of the state of the art disclosed above and for this purpose create an entirely new type of apparatus and method for depicting at least partly continuously the relative change in the functioning state of a brain, or for measuring the variation of the level.

The invention is based on determining from a signal measuring the functioning state of the brain, such as an EEG signal, the ratio between the activities of at least two frequency ranges. The activity is preferably a variable proportional to an amount of a specific rhythmic activity, such as spectral power, wavelet transformation, an entropy signal in AR (auto-regressive) models, or an activity shown by PCA (Principal component analysis), ICA- (Independent component analysis) or by a neural-network method

(Muthuswamy & Thakor, 1998; Samar, Bopardikar, Rao, & Swartz, 1999; Viertio-Oja, et al. 2004; Jackson & Sherratt, 2004; Stone, 2002; Robert, Gaudy, & Limoge, 2002; Vuckovic, Radivojevic, Chen, & Popovic, 2002).

In connection with one preferred embodiment of the invention, an index, which is sensitive to changes in the functioning state of the brain, describing the ratio between the spectral powers of the theta band of the Fz channel and the alpha band of the Pz channel is formed. In a second preferred embodiment of the invention, the ratio of the activity of the theta band to the activity of the alpha band is formed from one channel, for example the Fz channel.

hi a third preferred embodiment of the invention, the maxima of the spectral power of the different functioning states of the brain is detected and an index describing the activity and/or loading of the brain is formed from the ratios of the powers around the found maximum powers. If maxima are found, the activity of the maxima can be examined over very narrow frequency bands in the vicinity of the maximum frequency, thus obtaining a greater sensitivity for the measurement. This embodiment can also be applied to the individual calibration of the index (Klimesch, 1999).

hi a fourth preferred embodiment of the invention, the activity (quantity and quality of rhythmic activity) of the brain is first determined in different functioning states of the brain. After this, the activity obtained in the measuring situation is compared, for example with correlation, to these predefined activities of various functioning states.

In a fifth preferred, embodiment of the invention, maxima can be determined for a desired frequency range and ratios describing the activity of the brain are formed by the bands defined by the mean frequencies of these maxima.

More specifically, the apparatus according to the invention is characterized by what is stated in the characterizing portion of Claim 1.

The method according to the invention is, for its part, characterized by what is stated in the characterizing portion of Claim 13.

Considerable advantages are gained with the aid of the invention.

The invention provides a rapid, continuous, objective, simple, and sensitive way to measure and detect the functioning state of a brain. With the aid of the invention, it is possible to implement a portable measuring device for measuring the functioning states of brains.

It is possible to develop a method, by means of which cognitive workload can be measured in real time. Practical applications could be various methods for warning of a detrimental workload level (e.g., in monitoring and traffic tasks) and adaptive automation applications (the system takes into account a person's individual performance capability).

Figures and terminology

Figure 1 shows graphically an electroencephalograph measurement result, which can be applied to the invention. The frequency ranges are itemized with the aid of numbers: 20 delta, 21 theta, 22 alpha, and 23 beta.

Figure 2 shows graphically measurement results like those of Figure 1, which can be applied to solutions according to the invention.

Figure 3 shows a measuring arrangement according to the invention.

Figure 4 shows a flow diagram of the method according to the invention.

Figure 5 shows graphically traditional indices, in such a way that the beta/(theta+alpha) indices are shown on the left and the beta/alpha on the right. The test participants performed in turn one (single), two (dual), or four (multi) tasks simultaneously. On the basis of the repeated-measures analysis of variance, it is possible to observe that the traditional indices do not distinguish between different situations (beta/(theta+alpha) F(1.3, 25.5) = 0.275, p = 0.673 and beta/alpha F(1.3, 24.2) = 1.328, p = 0.271).

Figure 6 shows graphically indices according to the invention theta Fz/alpha Pz (on the left) and theta Fz/alpha Fz (on the right). The test participants performed in turn one (single), two (dual), or four (multi) tasks simultaneously. When measuring the index from the ratio of the Fz and Pz channels, an increase in the difficulty of the task increases the index according to the invention statistically significantly, F(1.4, 27.0) = 6.3, p = 0.011. hi addition, the Dual situation differs significantly from the Single situation (p = 0.008) and the Multi situation from the Dual situation (p = 0.029). Also, when using only the Fz channel, a statistical difference between the situations is observed (F(1.2, 25.0) = 8.6, p = 0.004). Also in this case, the various situations differ from each other statistically significantly (Single vs. Dual p = 0.002, Dual vs. Multi p = 0.014).

Figure 7 shows graphically the amplitude of the P300 component. A lower amplitude represents a higher workload. The test participants performed in turn one (single), two (dual), or four (multi) tasks simultaneously.

Figure 8 shows graphically the behaviour of the index according to the invention in the course of the day, when the test participants had slept for two or eight hours during the previous night. By repeated-measures analysis of variance, it can be observed that an increase in sleep pressure affects the index statistically significantly after both sleep deprivation (F(2.3, 34.0) = 6.2, p = 0.004) and normal sleep (F(2.4, 35.6) = 7.1, p = 0.002).

Figure 9 shows schematically the operation of one device according to the invention.

Figure 10 shows a block diagram of one apparatus according to the invention.

Figure 11 shows a block diagram of a second apparatus according to the invention.

In the present application, the following terminology, among others, is used in connection with the following reference numbers:

Examples

In the testing of the operation of the invention, measurement data collected in the Brain and Work Research Centre was used as material.

Workload

In the first part, various cognitive activities of the test participants, such as processes participating in attentiveness and working memory, were loaded with different degrees of difficulty. The degrees of difficulty were altered, for example, in such a way that the test participants performed one (single), two (dual), or four (multi) tasks simultaneously.

Frequency spectra, of which an example is in Figures 1 and 2, were first produced from the material.

Figure 1 shows in greater detail one preferred embodiment of the invention, in which the maxima of the amount of rhythmic activity are detected. According to Figure 1, the maxima 25 are detected in the desired frequency range. From the maxima, the mean frequency f_c is determined, as well as a suitable frequency band 2*Δf around the mean frequency. In this alternative, it is possible to compare, for example, the data (amplitude or spectral power) of two maxima, or alternatively to compare the properties of the maximum and/or the properties of the band relating to it with a fixed reference band (e.g., alpha 22). In addition, the sensitivity of the method can be improved at least by using individually defined frequency bands (Klimesch, 1999), with the removal of the trend from the signal as well as the removal of the envelope from the spectra.

The changes in the frequency bands were examined visually. It could be seen from the spectra that the EEG power ranges behave in different ways in different brain areas as the difficulty of the task changes. When the load of the tasks increases, the greatest changes could be detected in the theta of the frontal parts 21 and in the alpha of the posterior parts 22, according to Figures 1 and 2. Thus the study concentrated particularly on these channels and frequency ranges and on calculating the ratios of the various power ranges of the EEG between the different brain areas and on investigating the changes particularly in the frontal parts of the brain.

Figure 2 shows an example of the change in the frequency bands during different tasks. It can be seen from the figure that, among other things, as the load increases (one -> two -> four tasks simultaneously), the power of the theta band 21 increases in the Fz channel and the power of the alpha band 22 decreases in the Pz channel. A corresponding phenomenon has also been reported earlier (Gevins et al., 1998, Klimesch 1999).

According to the invention, a new index was formed. Based on the examinations of the frequency bands (as the workload increases, theta increases in Fz and alpha decreases in Pz) a new index is calculated; thetaFz/alphaPz. In the investigatory work for the invention, a study was also made as to whether the same information could be obtained by calculating the same ratio for a single channel thetaFz/alphaFz.

The left-hand side of Figure 5 shows the traditional indices beta/(theta+alpha) and on the right-hand side beta/alpha. The test participants performed in turn one, two, or four tasks simultaneously. On the basis of repeated-measures analysis of variance, it is possible to observe that the traditional indices do not distinguish different situations (beta/(theta+alpha) F(1.3, 25.5) = 0.275, p = 0.673 and beta/alpha F(1.3, 24.2) = 1.328, p = 0.271.

Figure 6 shows the indices theta Fz/alpha Pz (on the left) and theta Fz/alpha Fz (on the right). The test participants made in turn one, two, or four tasks simultaneously. When measuring the index from the ratio of the channels Fz and Pz, increasing the difficulty of the task increases the index according to the invention statistically significantly, F(1.4, 27.0) = 6.3, p = 0.011. In addition, the Dual situation differs significantly from the Single situation (p = 0.008) and me Multi situation from the Dual situation (p = 0.029). Also, when using only the Fz channel, a statistical difference is observed between the situations (F(1.2, 25.0) = 8.6, p = 0.004). In this case as well, the different situations differ from each other statistically significantly (Single vs. Dual p = 0.002, Dual vs. Multi p = 0.014).

The index according to the invention reacts more sensitively than traditional indices particularly to a change in the type of tasks and a change in the number of simultaneous tasks. Tasks that demand only vigilance receive the lowest values. Tasks that demand more processing, both by themselves and together with tasks demanding vigilance receive the next highest values. The highest index values of all are received by a multi-task of three different workload levels (4 simultaneous tasks).

Confirming the result using a second method

The result has been confirmed using also a second physiological method (event-related potential, ERP), according to Figure 7. It should be noted that the P300 analyses can only be performed for measurements, in which an auditive task has been included.

Figure 7 shows the amplitude of the P300 component. A lower amplitude indicates a higher workload. The test participants performed one, two, or four tasks simultaneously. Using repeated-measures analysis of variance it is observed that the amplitude of the P300 component decreases as the workload increases, F(1.5, 23.4) = 28.2, p = 0.001).

Sleep pressure

From literature, it is known that sleep deprivation causes an increase in theta activity, particularly in the frontal-lobe area (Cajochen, Foy, and Dijk, 1999; Cajochen et al. 2001). Thus sleep pressure causes changes of a similar direction to workload. Figure 8 shows the behaviour of the index according to the invention during a day, after the test participants have slept two or eight hours during the previous night, as well as after recovery sleep. The index increases during the day as the sleep pressure increases and returns during recovery sleep. The phenomenon can be seen both in sleep debt and after normal sleep. Using repeated-measures analysis of variance, it is possible to observe that the increase in sleep pressure effects the index statistically significantly both after sleep debt (F(2.3, 34.0) = 6.2, p = 0.004) and after normal sleep (F(2.4, 35.6) = 7.1, p = 0.002). During sleep debt, the index is greater than after normal sleep.

Depth of sleep

On the basis of literature, it is also known that theta (and delta) activity increases in all areas of the brain as the depth of sleep increases, whereas the amount of alpha activity decreases, especially in the posterior parts of the brain (Klimesch, 1999). Thus by placing the measuring sensors suitably, it is possible to use the method and the device also to estimate the depth of sleep.

The aforementioned examples show that the value of the index can describe factors (such as activeness, vigilance, sleep pressure, medication, alcohol, drugs) influencing the functioning state of the brain, as well as their combined effect.

Operation and measuring device

Measuring device

The calculation of the index according to the invention from the area of the frontal lobe makes a device possible, in which there is only one sensor. The area of the frontal lobe (especially the area of the forehead) is a favourable location for attaching the sensor, as the forehead is not covered by hair. In addition, it is also easy to place the sensor on the forehead by oneself.

The reduction in size of the measuring apparatuses and their cordless form permit also measuring and analysis during a work task, as measurement can take place without disturbing the workplace and task. The applicability of the method is thus diversified. The packaging of the measuring and analysis of the methods in a single small measuring device would open entirely new opportunities for research into physiological states.

Measuring arrangement

Figure 3 shows the measuring arrangement according to the invention. The figure shows top and side views of a person's head 5. hi the top view, the ears 6 appear on both sides of the head while the nose 7 is uppermost in the figure. One sensor, typically an electrode 1, is located preferably on the centre line F_z of the frontal lobe 50, though the location can, in principle be at any point of the frontal lobe 50, but the best results have been obtained in the front part of the centre line of the frontal lobe, and a second sensor 2 is on the centre line P_z of the parietal lobe 51. Similarly, in this case the location shown in the figure has given the best results so far, but the entire area of the parietal lobe 51 can probably be used. From the electrodes, the signal is transferred to the EEG measuring device 3 and forwarded from it to the signal-interpretation apparatus 4. The apparatus can be a laboratory apparatus, but the device according to the invention can be integrated to form a small, portable device, to which a measuring band containing the electrodes 1 and 2 is connected, either by wires or wirelessly.

The signal-interpretation device 4 should typically be able to calculate spectral powers (more generally, rhythmic variations in activity) on predefined frequency bands and to create ratios of their powers.

According to Figure 4, in the method according to the invention the following stages are performed:

measurement of brain signals 10, pre-processing of the measured signals 11, division into frequency ranges or search for maxima 12, comparison 13 of the properties of the frequency ranges or the maxima, and creation of an index, or making of an estimate on the basis of the comparison 14. Figure 9 shows schematically the operating principles of the device. There is at least one (possibly several) measuring points in the measuring device, from which a measured signal 30 is obtained. The device distinguishes the rhythmic components 31 from the signals of the measuring points 30 and, on the basis of these, calculates the index 33 by forming the ratio 32 of the rhythmic components.

The device according to the invention measures the index at least partly continuously, which in the present document refers, among other things, to the fact that the measurement data is formed from measurement data of seconds - minutes - hours and from this, on the basis of the measurement data information is created on the change and its direction.

hi other words, continuous measurement is used to form time series from individual measurement results, so that on the basis of the time series it is possible to determine the temporal behaviour of the measurement variable, for example with the aid of trends.

The index can be shown on the display device 34 and recorded 35 in parallel. The signal can also be further processed 36 and/or forwarded over a data link 37 for display, recording, or processing. The operation, particularly analysis can be performed simultaneously with the measurement (on-line), or afterwards (off-line).

The device can show numerically or graphically the momentary or cumulative value of the index, its trend graph, or some other variable calculated on the basis of these. The device can also produce a sound or light alarm, or some other signal when a defined limit value is exceeded / not reached.

Figure 10 shows one apparatus according to the invention, in which the measuring device 40 transfers the measuring data to a transmitter 41, which forwards the measuring data over a transfer path 42 to a receiver 43 and from there on to a data-processing unit 44. The transfer path 42 can be a conductor (galvanic or optical) as well as wireless (optical or radio-frequency).

According to Figure 11, the data-processing unit 44 can also be located in connection with the measuring device 40. hi the cases of both Figure 10 and Figure 11, the apparatus comprises a user interface and a display 45 in a suitable location in the system, for example, in connection with the data-processing unit (Figure 10), or in connection with the receiver (Figure 11).

hi the present application, the term ratio of the content of two frequency bands refers to the fact that the information of one frequency band is in the numerator of the graph and of the other in the denominator. Thus, for example, the ratio of the bands A and B refers to, among others, the following mathematical expressions, in which C represents other information:

A/B

B/A

(A+C)/B B/(A+C)

(B+C)/A

A/(B+C)

A/B + C

B/A + C

Applications for the invention

Indices according to the invention can be envisaged as being used in various applications, such as

1) to compare two different interfaces/work tasks, which has a higher loading effect physically and which cognitively,

2) the study of diseases and the efficacy of their treatment,

3) the investigation of the effects of chemical substances,

4) the investigation of factors due to an individual, such as age, learning, and intellectual as well as mental performance,

5) measurements during the workday can be used to determine objectively the loading of a workday and variations in workload,

6) in safety-critical work, the method could be used to send an alarm of excessive levels of loading and/or sleep pressure,

7) the index according to the invention can be combined with other methods measuring physiological state, for example, measurements of pulse or blood pressure,

8) the method can be used to detect the optimal physiological state (flow) for performing and learning a task, or to detect the stage of learning,

9) the method can also be used as part of the evaluation of the well-being of a person persons and animals.

The following are more detailed descriptions of some applications and service products.

Workload ergometer

Cognitive workload can be estimated in the same way as the loading of a person's autonomic nervous system, heart and circulatory organs, and lungs are estimated with the use of a bicycle ergometer. By varying the cognitive load in laboratory conditions by differently loading cognitive tests and by measuring the reactions of the electrical activity of the brain (workload of the central nervous system), as well as, for example, measuring the loading of the autonomic nervous system from an ECG. Thus it is possible to study the connection between cognitive workload and e.g. heart and circulatory diseases.

Loading effect of a Workday

The state of a person's central and autonomous nervous systems is tested by laboratory measurements (workload ergometer and bicycle ergometer), by means of which individual reference values for low, high, and optimal workload are obtained. After this, the person is measured during the workday. As a result, an estimate of the total workload and variation in the loading effect of the workday is obtained, when the results are compared with the baseline levels obtained previously from the person (work simulation). The tests can also be expanded into group and population studies.

Workload dosimeter

hi terms of product development, modern technology would permit the construction of a small, portable device, which could accompany employees in the same way as dosimeters used in radiation protection. Persons employed in work subject to radiation keep personal dosimeters with them during the workday, which are then sent for analysis to the Radiation Safety Institute. In the same way, people could keep a device measuring mental workload with them, which would be analysed at regular intervals by an authorized entity. If die workload exceeds critical limits, a person could be sent on workload leave, in the same way as one who has received excessive radiation is sent on radiation leave.

References

Fairclough, S. H., & Venables, L. (2006). Prediction of subjective states from psychophysiology: A multivariete approach. Biological Psychology, 71, 100-110.

Gevins, A., Leong, H., Du, R., Smith, M.E., Le, J., DuRousseau, D., Zhang, J., & Libove, J. (1995). Towards measurement of brain function in operational environments. Biological Psychology, 40, 169-186.

Gevins, A., Smith, M.E., Leong, H., McEvoy, L., Whitfield, S., Du, R., & Rush, G. (1998). Monitoring working memory load during computer-based tasks with EEG pattern recognition methods. Human Factors, 40( 1 ), 79-91.

Jackson, C, & Sherratt, M. (2004). A novel spatio-temporal decomposition of the EEG: derivation, validation and clinical application. Clin Neurophysiol, 7/5(1), 227-37. Kok, A. (1997). Event-related-potential (ERP) reflections of mental resources: a review and synthesis. Biological Psychology, 45, 19-56.

Klimesch, W. (1999). EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Brain Res Rev, 29(2-3), 169-95. Muthuswamy, J., & Thakor, N. V. (1998). Spectral analysis methods for neurological signals. J Neurosci Methods, 83(1), 1-14

Pope, A.T., Bogart, E.H., & Bartolome, D.S. (1995). Biocybernetic system evaluates indices of operator engagement in automated task. Biological Psychology, 40(1-2), 187- 195.

Robert, C, Gaudy, J.F., & Limoge, A. (2002). Electroencephalogram processing using neural networks. Clin Neurophysiol, 113(5), 694-701.

Samar, V.J., Bopardikar, A., Rao, R., & Swartz, K. (1999). Wavelet analysis of neuroelectric waveforms: a conceptual tutorial. Brain Lang, 66(1), 7-60. Sauseng, P., Klimesch, W., Schabus, M., & Doppelmayr, M. (2005). Frontoparietal EEG coherence in theta and upper alpha reflect central executive functions of working memory. International journal of psychophysiology, 57, 97-103.

Scerbo, M. W., Freeman, F.G., & Mikulka, PJ. (2003). A brain-based system for adaptive automation. Theoretical Issues in Ergonomic Sciences, 4(1-2), 200-219.

Slobounov, S.M., Fukada, K., Simon, R., Rearick, M., & Ray, W. (2000). Neurophysiological and behavioral indices of time pressure effects on visuomotor task performance. Cognitive brain research, 9, 287-298.

Stone, J.V. (2002). Independent component analysis: an introduction. Trends Cogn Sci, 6(2), 59-64.

Viertio-Oja, H., Maja, V., Sarkela, M., Talja, P., Tenkanen, N., Tolvanen-Laakso, H., Paloheimo, M., Vakkuri, A., Yli-Hankala, A., & Merilainen, P. (2004). Description of the Entropy algorithm as applied in the Datex-Ohmeda S/5 Entropy Module. Acta Anaesthesiol Scand, 48(2), 154-61.

Wilson, G.E., & Fisher, F. (1995). Cognitive task classification based upon topographic EEG data. Biological Psychology, 40, 239-250.

Vuckovic, A., Radivojevic, V., Chen, A.C., & Popovic, D. (2002). Automatic recognition of alertness and drowsiness from EEG by an artificial neural network. Med Eng Phys, 24(5), 349-60.

Claims

Claims:

1. Apparatus for determining the functioning state of a person's brain, which apparatus comprises, - at least one sensor (1, 2), which can be attached to the head (5) in order to measure the functioning state of the brain, using at least two frequency bands (alpha 22, theta 21),

- signal-processing means (3) for processing the signals measured by the sensor (1. 2), and - signal-interpretation means (4) for interpreting the measured signal, characterized in that it comprises, in addition,

- means (4) for detecting from the signal features indicating the functioning state of the brain, on the basis of at least two separate frequency bands (21, 22), as at least a partly continuous process, and - means (4) for forming a ratio of the information of these two frequency bands

(21, 22) as at least a partly continuous process.

2. Apparatus according to Claim 1, characterized in that it comprises means for determining variables describing the functioning state of the brain from the theta band (21) and the alpha band (22), as well as means for determining the ratio of these variables for forming an index.

3. Apparatus according to Claim 1 or 2, characterized in that it comprises means for measuring at least the theta band (21) from the area of the frontal lobe (50).

4. Apparatus according to Claim 3, characterized in that it comprises means for measuring also the alpha band (22) from the area of the frontal lobe (50).

5. Apparatus according to any of Claims 1 - 3, characterized in that it comprises means for measuring the theta band (21) from the area of the frontal lobe (50) and for measuring the alpha band (22) from the area of the parietal lobe (51).

6. Apparatus according to any of the above Claims, characterized in that it comprises means (4, 44) for determining the frequency bands (21, 22), for example, on the basis of the signal maxima.

7. Apparatus according to any of the above Claims, characterized in that it comprises means (4) for determining the functioning state of a brain, with the aid of activity found using a method depicting the amount of rhythmic variation, such as a spectral power, wavelet transform, entropy-signal, AR (auto-regressive) models or an activity shown by PCA- (Principal component analysis), ICA- (Independent component analysis) or by a neural-network method.

8. Apparatus according to any of the above Claims, characterized in that it comprises at least a frontal lobe sensor (1).

9. Apparatus according to any of the above Claims, characterized in that it comprises both a frontal lobe sensor (1) and a parietal-lobe sensor (2).

10. Apparatus according to any of the above Claims, characterized in that the measuring device is an EEG measuring device.

11. Apparatus according to any of Claims 1 - 8, characterized in that the measuring device is an NIRS or MEG measuring device.

12. Apparatus according to any of Claims 1 - 8, characterized in that the measuring device comprises means for processing data simultaneously with the measurement.

13. Method for determining the functioning state of a person's brain, in which method,

- the functioning state of the brain is measured using at least two frequency bands (alpha 22, theta 21),

- the signals measured are processed, and - the processed signal is interpreted, characterized in that,

- features indicating the functioning state of the brain are detected from the signal using at least two separate frequency bands (21, 22), as at least a partly continuous process, and

- a ratio of the information of these two frequency bands (21, 22) is formed as at least a partly continuous process.

14. Method according to Claim 13, characterized in that in it variables describing the functioning state of the theta band (21) and the alpha band (22) of the brain are determined, and the ratio of these variables is determined for forming an index.

15. Method according to Claim 13 or 14, characterized in that at least the theta band (21) is measured from the area of the frontal lobe (50).

16. Method according to Claim 15, characterized in that the alpha band (22) is also measured from the area of the frontal lobe (50).

17. Method according to any of Claims 13 - 16, characterized in that in it the theta band (21) is measured from the area of the frontal lobe (50) and the alpha band (22) is measured from the area of the parietal lobe (51).

18. Method according to any of the above Claims, characterized in that the frequency bands (21, 22) are defined, for example, on the basis of the signal maxima.

19. Method according to any of the above Claims, characterized in that the functioning state of the brain is determined with the aid of a method describing the amount of rhythmic variation, such as a spectral power, wavelet transform, entropy-signal, AR (auto-regressive) models or an activity shown by PCA- (Principal component analysis), ICA- (Independent component analysis) or by a neural-network method.

20. Method according to any of the above Claims, characterized in that in it at least a frontal lobe sensor (1) is used.

21. Method according to any of the above Claims, characterized in that in it both a frontal lobe sensor (1) and a parietal lobe sensor (2) are used.

22. Method according to any of the above Claims, characterized in that an EEG measuring device is used as the measuring device.

23. Method according to any of Claims 13 - 21, characterized in that an NIRS or MEG measuring device is used as the measuring device.

24. Method according to any of Claims 13 - 23, characterized in that the analysis is performed simultaneously with the measurement.

25. Method according to any of Claims 13 - 24, characterized in that a momentary or cumulative value, a trend graph, or some other variable calculated on their basis is formed.

26. Method according to any of the above Claims for a therapy.

27. Computer program product, which implements a method according to any of the above Claims.